kubeflow for poets

原文链接

K8s

Worker Node(s)

• kube-proxy: it is the Kubernetes network proxy that runs on each worker node. It listens to the api server and forward requests to the appropriate pod. Important for load-balancing.
• pod(s): consists of one or more containers that share network and storage resources as well as container runtime instructions. Pods are the smallest deployable unit in Kubernetes.

Each component usually includes two parts:

• Client code: The code that talks to endpoints to submit jobs. For example, the code to connect with the Google Cloud Machine Learning Engine.
• Runtime code: The code that does the actual job and usually runs in the cluster. For example, the code that prepares the model for training on Cloud MLE.

A component consists of an interface (inputs/outputs), the implementation (a Docker container image and command-line arguments) and metadata (name, description).